How I Build Cinematic Instagram Carousels
Connecting video and image generation platforms directly into LLMs via MCP allows AI filmmakers to automate visual storyboarding and production assets inside a single conversation interface.
What this lesson covers
DGI Kaos demonstrates how to automate cinematic Instagram carousels inside Claude by connecting Higgsfield through Model Context Protocol (MCP). By using Higgsfield's Soul Cinema model instead of default options, creators can generate styled multi-slide visual posts directly from single prompts within Claude.
Key takeaways from the creator
AI-extracted notes, not independently verified product claims. Timestamp links let you check each point in the original video.
- 02:27 ↗
Selecting specialized models like Soul Cinema in Higgsfield avoids the over-processed, artificial look common in standard web generators.
- 02:48 ↗
Connecting Higgsfield to Claude via MCP allows image and video creation directly inside a chat session without manual browser tab switching.
- 03:52 ↗
Prompting Claude with explicit structural rules enables the creation of reusable skills that automate multi-slide visual story builds.
- 05:03 ↗
Model selection within Claude affects execution speed and intelligence, ranging from utility tiers like Haiku up to advanced tiers like Fable 5 and Opus.
- 06:55 ↗
Isolating skill creation in a master chat session prevents context bloat when executing actual asset generation runs in clean child sessions.
Workflow outlined in the video
- Open Claude settings, navigate to Customize > Connectors, and click to add a custom connector.
- Copy and paste the Higgsfield MCP URL, save the connector, and authorize your Higgsfield account connection.
- Prompt Claude with your structural intent, defining slide counts, narrative flow, and aspect ratio rules to build a reusable carousel skill.
- Explicitly mandate the 'Soul Cinema' image model in your skill prompt instructions to ensure cinematic image outputs.
- Save the skill in a master session, then open a fresh chat session to trigger automated carousel production runs.
Before you use this workflow
These notes describe the source video at its publication date. Model access, pricing, connectors and interfaces may have changed. Check the original source and the provider’s current documentation before installing an add-on, connecting an account or spending credits.
ReelStack has not independently tested this workflow. Preview one representative shot and check motion, continuity and output quality before applying it to a full production. No result, cost saving or model capability is guaranteed.
Explore the tools
How to interpret the numbers
435 views captured 27 September 2026. ReelStack recommendation score: 16/100. Formula: radar-v5. This is ReelStack’s calculation, not a YouTube rating or a measure of factual accuracy.
Show the calculation
- 45% performance: 0.71× views vs 4 other channel videos observed at a similar age. View-evidence factor 30% (views / (views + 1,000))
- 25% freshness: 6/100 with a 14-day half-life
- 20% momentum: 2 views/day vs 4 channel baseline (4 comparable uploads), with the same view-evidence factor
- 10% engagement: 73/100; likes + 4× comments, smoothed with a 500-view neutral prior